Financial product recommendation method and system

Abstract

The invention provides a financial product recommendation method and system, and the method comprises the steps: obtaining the evaluation information of a user for a financial product, and determining a product vector based on the evaluation information; determining a user vector based on the evaluation information; determining an evaluation user set based on the evaluation information; calculating a financial product score based on the product vector, the user vector and the evaluation user set; the financial products to be recommended in the product set to be recommended are arranged in a descending order on the basis of the financial product scores to obtain an evaluation product set, and the financial products to be recommended are sequentially recommended to the users according to the sequential relation in the evaluation product set. The user preference can be fully identified, misunderstanding of the real user preference is avoided, and the matching degree of the recommended financial product and the user is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data recommendation, and particularly relates to a financial product recommendation method and system. Background Art

[0002] For some financial software, according to the needs of users, users can usually purchase some financial products on the financial software, such as funds, wealth management products, insurance, and so on.

[0003] At the same time, for existing financial software, it usually recommends suitable financial products according to the user's preferences. In the prior art, the recommendation algorithm usually simply recommends corresponding financial products based on the user's evaluation, but there is a problem that the user's preferences cannot be accurately identified, resulting in the recommended financial products not being accurately matched with the user's preferences. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a financial product recommendation method and system for solving the technical problems in the prior art.

[0005] On the one hand, the present invention provides the following technical solution. A financial product recommendation method includes: Obtaining evaluation information of the user on financial products, and determining a product vector based on the evaluation information; Determining a user vector based on the evaluation information; Determining an evaluation user set based on the evaluation information; Calculating a financial product score based on the product vector, the user vector, and the evaluation user set; Sorting the to-be-recommended financial products in the to-be-recommended product set in descending order based on the financial product score to obtain an evaluation product set, and sequentially recommending the to-be-recommended financial products to the user according to the order relationship in the evaluation product set.

[0006] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention first obtains the evaluation information of the user on financial products and determines a product vector based on the evaluation information; then determines a user vector based on the evaluation information; then determines an evaluation user set based on the evaluation information; then calculates a financial product score based on the product vector, the user vector, and the evaluation user set; finally, sorts the to-be-recommended financial products in the to-be-recommended product set in descending order based on the financial product score to obtain an evaluation product set, and sequentially recommends the to-be-recommended financial products to the user according to the order relationship in the evaluation product set. The present invention fully considers the relationship between users and the relationship between users and products, can fully identify the user's preferences, avoid misunderstanding of the real user preferences, and improve the matching degree between the recommended financial products and the user.

[0007] Preferably, the step of determining the product vector based on the evaluation information includes: Calculating a product concept score based on the first evaluation information : ; In the formula, represents the score given by the -th user to the -th financial product, represents the average score of the financial products evaluated by the -th user; Obtaining the product score set of the financial products evaluated by the -th user from the first evaluation information, converting the product score set into a product score vector, and subtracting the product concept score from each element in the product score vector to obtain a first score vector ; Calculating a product interaction quantity based on the first score vector : : ; In the formula, is a multi-layer perceptron, represents the embedding of the -th user, is a concatenation operation; Determining a product vector based on the product interaction quantity : : ; ; ; In the formula, , , respectively represent a first activation function, a second activation function, and a user aggregation function, is a product attention weight, represents the embedding of the -th financial product, represents the set of users who have evaluated the -th financial product, , , are respectively the first, second, and third weight vectors, respectively represent a first bias, a second bias, and a third bias.

[0008] Preferably, the step of determining the user vector based on the evaluation information includes: Calculate the user concept score based on the first evaluation information : ; In the formula, represents the score of the th user for the th financial product, represents the average score of all users for the th financial product; Obtain the user score set of all users for the th financial product from the first evaluation information, convert the user score set into a user score vector, and subtract the user concept score from each element in the user score vector to obtain a second score vector ; Calculate the user interaction volume based on the second score vector : : ; In the formula, is a multi-layer perceptron, represents the embedding of the th financial product, is a concatenation operation; Determine the user vector based on the user interaction volume : : ; ; ; In the formula, , , respectively represent the first activation function, the second activation function, and the product aggregation function, is the user attention weight, represents the embedding of the th user, represents the set of financial products evaluated by the th user, , , are the first, second, and third weight vectors respectively, respectively represent the first bias, the second bias, and the third bias.

[0009] Preferably, the step of determining the evaluation user set based on the evaluation information includes: Calculate the user and the user Similarity between : ; In the formula, represents the user and the user The set of financial products jointly evaluated, represents The number of financial products in , represents the user , the user for The score of the th financial product in Determine the similarity between the user and all users in the remaining user set, and remove users with a similarity less than the preset similarity from the remaining user set to obtain the evaluation user set of the user .

[0010] Preferably, the step of calculating the financial product score based on the product vector, the user vector, and the evaluation user set includes: Based on the product vector , the user vector Calculate the user preference value : ; In the formula, , , Are the first, second, and third weight vectors respectively, Represents the first activation function, Represents the third deviation; Based on the user preference value Calculate the user product preference value : ; In the formula, Represents the fusion weight, Represents the evaluation user set, Represents the user And the similarity between the user and the th user in the evaluation user set, Represents the th user in the evaluation user set's preference value for the th financial product; Based on the user product preference value Calculate the financial product score : ; Wherein, represents the average score of all users for the th financial product, represents the average score of the th user for the financial product he / she evaluated.

[0011] In a second aspect, the present invention provides the following technical solution. A financial product recommendation system, the system includes: A first vector module, configured to obtain evaluation information of users for financial products, and determine a product vector based on the evaluation information; A second vector module, configured to determine a user vector based on the evaluation information; A set module, configured to determine an evaluation user set based on the evaluation information; A scoring module, configured to calculate a financial product score based on the product vector, the user vector, and the evaluation user set; A recommendation module, configured to sort the to-be-recommended financial products in the to-be-recommended product set in descending order based on the financial product score to obtain an evaluation product set, and sequentially recommend the to-be-recommended financial products to the user according to the order relationship in the evaluation product set.

[0012] Preferably, the first vector module includes: A product concept scoring sub-module, configured to calculate a product concept score based on the first evaluation information : ; Wherein, represents the score of the th user for the th financial product, represents the average score of the th user for the financial product he / she evaluated; A first scoring sub-module, configured to obtain a product score set of the th user for the financial product he / she evaluated from the first evaluation information, convert the product score set into a product score vector, and subtract each element in the product score vector by the product concept score to obtain a first score vector ; A product interaction quantum module, configured to calculate a product interaction quantity based on the first score vector : ; Wherein, is a multi-layer perceptron, represents the Embedding of a user is a splicing operation The product vector sub-module is used to determine the product vector based on the product interaction quantity : : ; ; ; In the formula , , respectively represent the first activation function, the second activation function, and the user aggregation function is the product attention weight represents the th embedding of the financial product represents the set of users who have evaluated the th financial product , , are the first, second, and third weight vectors respectively respectively represent the first deviation, the second deviation, and the third deviation

[0013] Preferably, the steps of the second vector module include The user concept scoring sub-module is used to calculate the user concept score based on the first evaluation information : ; In the formula represents the score of the th user for the th financial product represents the average score of all users for the th financial product The second scoring sub-module is used to obtain the set of user scores of all users for the th financial product from the first evaluation information, convert the set of user scores into a user score vector, and subtract the user concept score from each element in the user score vector to obtain the second score vector ; The user interaction quantity sub-module is used to calculate the user interaction quantity based on the second score vector : ; In the formula is a multi-layer perceptron represents the The embedding of a financial product is a splicing operation; The user vector sub-module is used to determine the user vector based on the user interaction volume : : ; ; ; In the formula, , , respectively represent the first activation function, the second activation function, and the product aggregation function, is the user attention weight, represents the embedding of the th user, represents the set of financial products evaluated by the th user, , , are the first, second, and third weight vectors respectively, respectively represent the first deviation, the second deviation, and the third deviation.

[0014] Thirdly, the present invention provides the following technical solution. A computer includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the financial product recommendation method as described above is implemented.

[0015] Fourthly, the present invention provides the following technical solution. A storage medium stores a computer program, and when the computer program is executed by a processor, the financial product recommendation method as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 is a flowchart of the financial product recommendation method provided in Embodiment 1 of the present invention; Figure 2 is a structural block diagram of the financial product recommendation system provided in Embodiment 2 of the present invention; Figure 3 is a schematic diagram of the hardware structure of a computer provided in another embodiment of the present invention.

[0018] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings. Specific embodiments

[0019] The embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the embodiments of the present invention and should not be construed as limiting the present invention.

[0020] In the description of the embodiments of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the embodiments of the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the present invention.

[0021] In addition, the terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present invention, "a plurality" means two or more unless otherwise specifically defined.

[0022] In the embodiments of the present invention, unless otherwise clearly defined and limited, the terms "installed", "connected", "connected", "fixed", etc. should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and may be the communication inside two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0023] Embodiment 1 In Embodiment 1 of the present invention, as Figure 1 shown, a financial product recommendation method includes: S1. Obtain the evaluation information of the user on the financial product, and determine the product vector based on the evaluation information; Among them, the evaluation information here specifically includes the evaluation of the financial product by the user, which can be specifically represented as the corresponding score, and the degree of preference of the user for the financial product is determined by the size of the score.

[0024] Among them, the step S1 includes: S11. Calculate the product concept score based on the first evaluation information : ; In the formula, represents the score given by the -th user to the -th financial product, and represents the average score of the financial products evaluated by the -th user.

[0025] S12. Obtain the product score set of the financial products evaluated by the -th user from the first evaluation information, convert the product score set into a product score vector, and subtract the product concept score from each element in the product score vector to obtain the first score vector .

[0026] S13. Calculate the product interaction volume based on the first score vector : : ; In the formula, is a multi-layer perceptron, represents the embedding of the -th user, and is a concatenation operation.

[0027] S14. Determine the product vector based on the product interaction volume : ; ; ; In the formula, , , respectively represent the first activation function, the second activation function, and the user aggregation function, is the product attention weight, represents the embedding of the -th financial product, represents the set of users who have evaluated the -th financial product, , , are the first, second, and third weight vectors respectively, and represent the first deviation, the second deviation, and the third deviation respectively; Specifically, the first, second, and third weight vectors, as well as the first bias, second bias, and third bias here, are the weights and biases in the graph neural network, and the first activation function is the hyperbolic tangent activation function. The second activation function is specifically an activation function, and the product vector here specifically reflects the influence of the characteristics of the financial product itself on user preferences.

[0028] S2. Determine the user vector based on the evaluation information; Among them, the step S2 includes: S21. Calculate the user concept score based on the first evaluation information : ; In the formula, represents the score of the th user for the th financial product, represents the average score of all users for the th financial product.

[0029] S22. Obtain the user score set of all users for the th financial product from the first evaluation information, convert the user score set into a user score vector, and subtract the user concept score from each element in the user score vector to obtain the second score vector .

[0030] S23. Calculate the user interaction volume based on the second score vector : ; In the formula, is a multi-layer perceptron, represents the embedding of the th financial product, is a concatenation operation.

[0031] S24. Determine the user vector based on the user interaction volume : ; ; ; In the formula, , , respectively represent the first activation function, the second activation function, and the product aggregation function, is the user attention weight, represents the embedding of the th user, represents the set of financial products evaluated by the th user, , , are the first, second, and third weight vectors respectively, represent the first deviation, second deviation, and third deviation respectively; Specifically, the process of determining the user vector here is similar to the process of determining the above product vector, except that the objects targeted are different, one is for users and the other is for financial products. The user vector here specifically reflects the influence of the characteristics of the user himself on the user's preferences.

[0032] S3. Determine the set of evaluated users based on the evaluation information; Among them, the step S3 includes: S31. Calculate the similarity between user and user : ; In the formula, represents the set of financial products jointly evaluated by user and user , represents the number of financial products in , represent the ratings of user , user on the th financial product in .

[0033] S32. Determine the similarity between user and all users in the remaining user set, and remove the users with similarity less than the preset similarity from the remaining user set to obtain the set of evaluated users of user ; Specifically, the similarity here specifically reflects the degree of association between users, which can be specifically reflected as the number of common financial products that users like or dislike each other.

[0034] S4. Calculate the financial product rating based on the product vector, the user vector, and the set of evaluated users; Among them, the step S4 includes: S41. Calculate the user preference value based on the product vector , the user vector ​ : ; Wherein, , , are the first, second, and third weight vectors respectively, represents the first activation function, represents the third deviation.

[0035] S42. Calculate the user's product preference value based on the user preference value : ; Wherein, represents the fusion weight, represents the set of evaluation users, represents the user and the th user in the set of evaluation users represents the similarity between the th user in the set of evaluation users and the th financial product,

[0036] S43. Calculate the financial product score based on the user's product preference value : ; Wherein, represents the average score of all users for the th financial product, represents the average score of the th user for the financial product they evaluated; Specifically, the above financial product score comprehensively considers the overall like or dislike tendency of the user, the liked or disliked degree of the financial product, and the interaction between the user and the financial product, so that the financial product score can more accurately reflect the true preference degree of the user for the financial product.

[0037] S5. Rank the financial products to be recommended in the set of products to be recommended in descending order based on the financial product score to obtain an evaluation product set, and recommend the financial products to be recommended to the user in sequence according to the order relationship in the evaluation product set; Specifically, rank the financial products to be recommended in descending order according to the financial product score, and then recommend the financial products to the corresponding users in sequence.

[0038] The financial product recommendation method provided in the first embodiment of the present invention first obtains the evaluation information of users on financial products, determines the product vector based on the evaluation information; then determines the user vector based on the evaluation information; then determines the evaluation user set based on the evaluation information; then calculates the financial product score based on the product vector, the user vector, and the evaluation user set; finally, sorts the to-be-recommended financial products in the to-be-recommended product set in descending order based on the financial product score to obtain the evaluated product set, and recommends the to-be-recommended financial products to the user in sequence according to the order relationship in the evaluated product set. The present invention fully considers the relationship between users and the relationship between users and products, can fully identify user preferences, avoid misunderstanding of real user preferences, and improve the matching degree between the recommended financial products and users.

[0039] Embodiment 2 As Figure 2 shown, Embodiment 2 of the present invention provides a financial product recommendation system, and the system includes: The first vector module 1 is used to obtain the evaluation information of users on financial products and determine the product vector based on the evaluation information; The second vector module 2 is used to determine the user vector based on the evaluation information; The set module 3 is used to determine the evaluation user set based on the evaluation information; The scoring module 4 is used to calculate the financial product score based on the product vector, the user vector, and the evaluation user set; The recommendation module 5 is used to sort the to-be-recommended financial products in the to-be-recommended product set in descending order based on the financial product score to obtain the evaluated product set, and recommend the to-be-recommended financial products to the user in sequence according to the order relationship in the evaluated product set.

[0040] The first vector module 1 includes: The product concept scoring sub-module is used to calculate the product concept score based on the first evaluation information : ; In the formula, represents the score of the rd user on the th financial product, represents the average score of the th user on the financial products he evaluates; The first scoring sub-module is used to obtain the product score set of the rd user on the financial products he evaluates from the first evaluation information, convert the product score set into a product score vector, and subtract the product concept score from each element in the product score vector to obtain the first score vector ; A product interaction quantum module, for calculating the product interaction quantity based on the first scoring vector : : ; In the formula, is a multi-layer perceptron, represents the embedding of the -th user, is a concatenation operation; A product vector quantum module, for determining the product vector based on the product interaction quantity : : ; ; ; In the formula, , , respectively represent the first activation function, the second activation function, and the user aggregation function, is the product attention weight, represents the embedding of the -th financial product, represents the set of users who have evaluated the -th financial product, , , are the first, second, and third weight vectors respectively, respectively represent the first deviation, the second deviation, and the third deviation.

[0041] The steps of the second vector module 2 include: A user concept scoring sub-module, for calculating the user concept score based on the first evaluation information : ; In the formula, represents the score of the -th user for the -th financial product, represents the average score of all users for the -th financial product; A second scoring sub-module, for obtaining the set of user scores of all users for the -th financial product from the first evaluation information, converting the set of user scores into a user score vector, and subtracting the user concept score from each element in the user score vector to obtain the second scoring vector ; A user interaction quantum module for calculating the user interaction amount based on the second scoring vector : : ; In the formula, is a multi-layer perceptron, represents the embedding of the th financial product, is a splicing operation; A user vector quantum module for determining the user vector based on the user interaction amount : : ; ; ; In the formula, , , respectively represent the first activation function, the second activation function, and the product aggregation function, is the user attention weight, represents the embedding of the th user, represents the set of financial products evaluated by the th user, , , are the first, second, and third weight vectors respectively, respectively represent the first bias, the second bias, and the third bias.

[0042] The set module 3 includes: A similarity sub-module for calculating the similarity between user and user : : ; In the formula, represents the set of financial products jointly evaluated by user and user , represents the number of financial products in , , represent the scores of user , user on the th financial product in ; A removal sub-module for determining user The similarity with all users in the remaining user set, and users with a similarity less than a preset similarity are excluded from the remaining user set to obtain an evaluation user set of users. of the evaluation user set.

[0043] The scoring module 4 includes: A preference sub-module for calculating a user preference value based on the product vector and the user vector : : ; In the formula, , , are the first, second, and third weight vectors respectively, represents the first activation function, represents the third deviation; A preference sub-module for calculating a user product preference value based on the user preference value : : ; In the formula, represents the fusion weight, represents the evaluation user set, represents the user and the similarity between the user and the th user in the evaluation user set, represents the th user in the evaluation user set's preference value for the th financial product; A financial product scoring sub-module for calculating a financial product score based on the user product preference value : ; In the formula, represents the average score of all users for the th financial product, represents the th user's average score for the financial product they evaluated.

[0044] In some other embodiments of the present invention, the present invention provides the following technical solution. A computer includes a memory 102, a processor 101, and a computer program stored on the memory 102 and executable on the processor 101. When the processor 101 executes the computer program, the financial product recommendation method described above is implemented.

[0045] Specifically, the above-mentioned processor 101 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured as one or more integrated circuits for implementing the embodiments of the present invention.

[0046] Among them, the memory 102 may include a mass storage for data or instructions. By way of example and not limitation, the memory 102 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In a suitable case, the memory 102 may include removable or non-removable (or fixed) media. In a suitable case, the memory 102 may be inside or outside the data processing device. In a specific embodiment, the memory 102 is a non-volatile memory. In a specific embodiment, the memory 102 includes a read-only memory (ROM) and a random access memory (RAM). In a suitable case, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or a flash memory, or a combination of two or more of these. In a suitable case, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM may be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0047] The memory 102 can be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 101.

[0048] The processor 101 reads and executes the computer program instructions stored in the memory 102 to implement the above-mentioned financial product recommendation method.

[0049] In some of the embodiments, the computer may further include a communication interface 103 and a bus 100. Among them, as Figure 3 shown, the processor 101, the memory 102, and the communication interface 103 are connected through the bus 100 and complete communication with each other.

[0050] The communication interface 103 is used to implement communication between the various modules, devices, units, and / or devices in the embodiments of the present invention. The communication interface 103 can also implement data communication with other components, such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations, etc.

[0051] The bus 100 includes hardware, software, or both, and couples components of a computer device to each other. The bus 100 includes, but is not limited to, at least one of the following: Data Bus, Address Bus, Control Bus, Expansion Bus, Local Bus. By way of example and not limitation, the bus 100 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable bus or a combination of two or more of these. In a suitable case, the bus 100 may include one or more buses. Although embodiments of the present invention describe and illustrate specific buses, the present invention contemplates any suitable bus or interconnect.

[0052] The computer may execute the financial product recommendation method of the present invention based on the obtained financial product recommendation system, thereby realizing financial product recommendation.

[0053] In still some other embodiments of the present invention, in combination with the above-mentioned financial product recommendation method, embodiments of the present invention provide the following technical solution: a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned financial product recommendation method is realized.

[0054] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0055] More specific examples (non-exhaustive list) of the readable medium include the following: an electrical connection part (electronic device) having one or more wirings, a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing as necessary, and then stored in a computer memory.

[0056] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0057] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity in description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as within the scope described in this specification.

[0058] The above-described embodiments merely represent several implementation manners of the present invention. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the appended claims.

Claims

1. A financial product recommendation method, characterized in that: include: Obtaining user evaluation information on financial products, and determining a product vector based on the evaluation information; determining a user vector based on the evaluation information; Determine a set of evaluation users based on the evaluation information; Calculate a financial product score based on the product vector, the user vector, and the evaluation user set; The financial products to be recommended in the product set to be recommended are arranged in descending order based on the financial product scores to obtain an evaluation product set, and the financial products to be recommended are recommended to the user in sequence according to the order in the evaluation product set.

2. The financial product recommendation method according to claim 1, characterized in that: The step of determining a product vector based on the evaluation information comprises: Calculate the product concept score based on the first evaluation information : ； In the formula, Indicates User to Ratings of financial products. Indicates The average rating of the financial products reviewed by users; Obtain the first evaluation information from the first evaluation information The product rating set of the financial products evaluated by each user is converted into a product rating vector, and the product concept score is subtracted from each element in the product rating vector. , to get the first score vector ; Based on the first scoring vector Calculate product interactions : ； In the formula, is a multi-layer perceptron, Indicates User embedding, For splicing operation; Based on the product interaction volume Determine the product vector : ； ； ； In the formula, , , Respectively represent the first activation function, the second activation function, and the user aggregation function, is the product attention weight, Indicates The embedding of financial products, Expressing the The set of users who have evaluated financial products. , , are the first, second, and third weight vectors respectively, They represent the first deviation, the second deviation, and the third deviation respectively.

3. The financial product recommendation method according to claim 1, characterized in that: The step of determining the user vector based on the evaluation information comprises: Calculate the user concept score based on the first evaluation information : ； In the formula, Indicates User to Ratings of financial products. Indicates all users' The average rating of financial products; Obtain all users' evaluations of the first A user rating set of financial products is converted into a user rating vector, and the user concept score is subtracted from each element in the user rating vector. , to obtain the second score vector ; Based on the second scoring vector Calculating user interactions : ； In the formula, is a multi-layer perceptron, Indicates The embedding of financial products, For splicing operation; Based on the user interaction Determine the user vector : ； ； ； In the formula, , , Represent the first activation function, the second activation function, and the product aggregation function respectively. is the user attention weight, Indicates User embedding, Indicates A collection of financial products rated by users, , , are the first, second and third weight vectors respectively, They represent the first deviation, the second deviation, and the third deviation respectively.

4. The financial product recommendation method according to claim 1, characterized in that: The step of determining the evaluation user set based on the evaluation information comprises: Counting users With users Similarity between : ； In the formula, Indicates user With users A collection of financial products that have been reviewed together. express The number of financial products in , Indicates user ,user right Middle Ratings of financial products; Identify users The similarity between the user and all users in the remaining user set is such that users with similarity less than a preset similarity are removed from the remaining user set to obtain the user The collection of users who reviewed .

5. The financial product recommendation method according to claim 1, characterized in that: The step of calculating the financial product score based on the product vector, the user vector, and the evaluation user set comprises: Based on the product vector , the user vector Calculate user preference value : ； In the formula, , , are the first, second and third weight vectors respectively, represents the first activation function, Indicates the third deviation; Based on the user preference value Calculate user product preference value : ； In the formula, represents the fusion weight, Represents the set of evaluation users. Indicates user and the evaluation user set The similarity between users, Indicates the first User to User preference value of each financial product; Based on the user product preference value Calculating financial product scores : ； In the formula, Indicates all users' The average rating of financial products, Indicates The average rating given by users to the financial products they reviewed.

6. A financial product recommendation system, characterized in that: The system comprises: A first vector module, configured to obtain user evaluation information on a financial product and determine a product vector based on the evaluation information; A second vector module, configured to determine a user vector based on the evaluation information; A collection module, used to determine a collection of evaluation users based on the evaluation information; A scoring module, used to calculate a score of a financial product based on the product vector, the user vector, and the evaluation user set; The recommendation module is used to arrange the financial products to be recommended in the product set to be recommended in descending order based on the financial product scores to obtain an evaluation product set, and recommend the financial products to be recommended to the user in sequence according to the order in the evaluation product set.

7. The financial product recommendation system according to claim 6, characterized in that: The first vector module comprises: A product concept scoring submodule is used to calculate a product concept score based on the first evaluation information. : ； In the formula, Indicates User to Ratings of financial products. Indicates The average rating of the financial products reviewed by users; The first scoring submodule is used to obtain the first evaluation information from the first evaluation information. The product rating set of the financial products evaluated by each user is converted into a product rating vector, and the product concept score is subtracted from each element in the product rating vector. , to get the first score vector ; A product interaction quantum module for based on the first scoring vector Calculate product interactions : ； In the formula, is a multi-layer perceptron, Indicates User embedding, For splicing operation; Product vector module for interaction based on the product Determine the product vector : ； ； ； In the formula, , , Respectively represent the first activation function, the second activation function, and the user aggregation function, is the product attention weight, Indicates The embedding of financial products, Indicates The set of users who have reviewed financial products. , , are the first, second and third weight vectors respectively, They represent the first deviation, the second deviation, and the third deviation respectively.

8. The financial product recommendation system according to claim 6, characterized in that: The steps of the second vector module include: A user concept scoring submodule is used to calculate the user concept score based on the first evaluation information. : ； In the formula, Indicates User to Ratings of financial products. Indicates all users' The average rating of financial products; The second scoring submodule is used to obtain all users' evaluations of the first evaluation information from the first evaluation information. A user rating set of financial products is converted into a user rating vector, and the user concept score is subtracted from each element in the user rating vector. , to obtain the second score vector ; A user interaction quantum module for based on the second scoring vector Calculating user interactions : ； In the formula, is a multi-layer perceptron, Indicates The embedding of financial products, For splicing operation; The user vector submodule is used to Determine the user vector : ； ； ； In the formula, , , Represent the first activation function, the second activation function, and the product aggregation function respectively. is the user attention weight, Indicates User embedding, Indicates A collection of financial products rated by users, , , are the first, second and third weight vectors respectively, They represent the first deviation, the second deviation, and the third deviation respectively.

9. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the financial product recommendation method according to any one of claims 1 to 5 is implemented.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the financial product recommendation method according to any one of claims 1 to 5 is implemented.

Images